Evaluation of Naïve Bayes and J48 Learning Algorithm for Text Classification

Text mining is one of the data mining applications. One of the importance text mining techniques is automatic text classification. For text classification many methods have been proposed. In this paper we implement and evaluate two text classification methods using naïve bayes and J48 algorithms under two term frequency and tfidf term weighting methods. According to results, the J48 method has better performance than naïve bayes in term frequency and tfidf term weighting methods. Also according to results tfidf has better performance than term frequency method in text classification. Selecting a good learning method and also term weighting method can improve final performance of text classification.